Cerebral ventricular enlargement as a generalized feature of schizophrenia: a distribution analysis on 502 subjects

Abstract

Enlargement of cerebral ventricles is one of the most replicated biological features, and the one quantitatively most deviant in schizophrenia. It occurs in the early phases of the disease and may have pathogenetic relevance. Whether this abnormality is limited to a specific subgroup of patients or is a common feature to most or all patients affected by schizophrenia, however, is still a matter of debate. The answer to this question would improve our comprehension of the nature of this abnormality and contribute to the debate between the competing hypotheses of biological homogeneity vs heterogeneity of schizophrenia. We performed a distribution analysis of lateral ventricular dimensions of 340 schizophrenic patients and 162 non-psychiatric controls. All subjects underwent cerebral computerized tomographic scan, and ventricular dimensions were expressed as ventricular brain ratio (VBR). After removing the effect of confounding variables (age, sex and type of scanner) on individual VBR, data were power-transformed and different distribution hypotheses were tested by means of the maximum log-likelihood ratio method. Our findings indicate that, in the mixed sample of patients and controls, a mixture of two gaussian curves represents the distribution better than a single gaussian curve, but no evidence emerged leading to rejection of the normality hypothesis in the schizophrenic patients sample. Lateral ventricular enlargement in schizophrenia is not a marker of a discrete subgroup of schizophrenia, but occurs in most, if not all, schizophrenic patients. This supports the hypothesis of biological homogeneity of the disease, at least relative to its major brain morphological abnormality.